Forum Discussion

AbiramiRajaram's avatar
AbiramiRajaram
Copper Contributor
Feb 10, 2025

Seeking Best Practices for Performance Testing Bots in Microsoft Teams

Teams Meeting Bot

This bot is automatically installed in all scheduled meetings across the organization. It interacts with the Microsoft Graph API via the 'api/messages' endpoint in the Bot Framework to retrieve meeting transcripts. These transcripts are then processed by an LLM model to generate a summarized version of the meeting.

Chatbot

This is a personal Teams chatbot built using the Bot Framework. It streams real-time responses from an LLM model based on user queries.

We are planning a performance test for these bots. What would be the standard procedure to achieve this?

Looking forward to your insights.

BOT
Chat
meetings
microsoft teams
microsoft teams bot
Performance Test
performance tools
  • AbiramiRajaram's avatar
    AbiramiRajaram
    Copper Contributor
    Feb 14, 2025

    Nivedipa-MSFT 

    Many Thanks for writing back quickly. 

    Can you suggest some automation tools & documentation to perform Load testing in Azure DevOps?

  • Nivedipa-MSFT's avatar
    Nivedipa-MSFT
    Icon for Microsoft rankMicrosoft
    Feb 11, 2025

    @AbiramiRajaram - Thank you for bringing this issue to our attention. To effectively conduct performance testing for your Teams Meeting Bot and Chatbot, here are some best practices and standard procedures you can follow:

    1. Define Performance Metrics: Identify key performance indicators (KPIs) such as response time, throughput, error rates, and resource utilization. This will help you measure the performance of your bots accurately.
    2. Simulate Real-World Scenarios: Create test cases that mimic real-world usage scenarios. For the Teams Meeting Bot, simulate various meeting sizes and durations. For the Chatbot, simulate different types of user queries and interactions.
    3. Load Testing: Use load testing tools to simulate multiple concurrent users interacting with the bots. This will help you understand how the bots perform under high load conditions.
    4. Stress Testing: Push the bots to their limits by gradually increasing the load until they fail. This will help you identify the breaking point and potential bottlenecks.
    5. Monitor and Analyze: Continuously monitor the performance metrics during the tests. Use tools like Azure Monitor or Application Insights to collect and analyze the data.
    6. Optimize and Iterate: Based on the test results, identify areas for improvement and optimize the bot's code and infrastructure. Repeat the tests to ensure the optimizations are effective.
    7. Test and Debug Locally: Before deploying the bots for performance testing, ensure they are thoroughly tested and debugged locally. This helps in identifying and fixing issues early on.
    8. Use Azure DevOps: Utilize Azure DevOps for managing the performance testing process. Create pipelines to automate the testing and reporting.
    9. Documentation and Best Practices: Refer to Microsoft's documentation and community forums for additional best practices and guidelines on performance testing bots in Microsoft Teams.

     

     

    Thanks, 

    Nivedipa 

    ------------------------------------------------------------------------------------------ 

    If the response is helpful, please click "**Mark as Best Response**" and like it. You can share your feedback via Microsoft Teams Developer Feedback link. Click here to escalate. 

     

Resources